Principal Investigators(Last, first, middle):KLEINFELD, DAVID and ROSEN, BRUCE R. Functional magnetic resonant imaging (fMRI) is the only means to infer neuronal activity within the entire volume of the human brain. A powerful aspect of fMRI concerns coordinated fluctuations in the amplitude of blood oxygen level dependent (BOLD) signals across distant regions of the brain, which are interpreted as resting-state functional connections. Here we address the underlying biophysical mechanism that underlies resting-state functional connectivity. Our hypothesis is that the natural ultra-low frequency oscillations in the smooth muscle of arteriole walls, termed vasomotion, acts as an intermediate oscillator that links oscillations in neuronal activity with the blood oxygenation and thus fMRI signals. Rodent models permit us to test this hypothesis through detailed two-photon imaging, advanced fMRI measurements, and manipulations of cortical vascular dynamics and blood oxygenation under controlled conditions. We then advance the spatial resolution of ultra high field MR imaging in humans to image single intracortical vessels, with 100 micrometer or better resolution, to test whether vasomotion may be a unifying mechanism for resting and task-driven fMRI signals. The results of these studies have two consequences. One is to provide the underpinnings for interpreting resting state connectivity relative to neuronal projections. The second is a new model of mapping functional connections via changes in arteriole volume. In particular, the strong homologies between the physiology of rodents and primates suggest that these methods can be extended to map resting- state functional connections in the human brain with higher resolution and greater precision than previously achieved. This new mechanistic insight will advance our use of fMRI to study cognition and a variety of neuropsychiatric disorders.